Wireless communication systems, i.e. systems that provide communication services to wireless communication devices such as mobile phones, smartphones etc. (often denoted by UE that is short for user equipment), have evolved during the last decade into systems that must utilize the radio spectrum in the most efficient manner possible. A reason for this is the ever increasing demand for high speed data communication capabilities in terms of, e.g., bitrate and to provide these capabilities at any given time, at any geographical location and also in scenarios where the wireless communication device is moving at a high speed, e.g., on board a high speed train.
To meet this demand, within the third generation partnership project, 3GPP, work is being done regarding possible enhancements to radio resource management, RRM, performance in high speed train environments. The justification is that there are railways such as Japan Tohoku Shinkansen (running at 320 km/h), German ICE (330 km/h), AGV Italo (400 km/h), and Shanghai Maglev (430 km/h) which vehicles travel at greater than 300 km/h and where there is demand for using mobile services. In a motivation contribution to 3GPP RAN#66, RP-141849, four scenarios of interest to wireless communication network operators are disclosed. In a number of these scenarios, there is a dedicated network to provide railway coverage of the cellular system; either as a standalone network, or used in conjunction with a public network which is not specifically designed to provide high speed train coverage. The four scenarios in RP-141849 can be summarized as follows:
Scenario 1: A dedicated network is deployed along the railways (such as antenna nodes in the form of remote radio head, RRH, deployments). Separate carriers are utilized for dedicated and public networks. By sharing the same Cell identity among multiple RRHs, handover success rate can be increased to some extent.
Scenario 2: Separate carriers are utilized for high speed scenario. One carrier with good coverage serves as a primary cell, PCell, for mobility management. One carrier at high frequency may provide the good data transmission. Carrier aggregation, CA, or dual connectivity, DC, could be applied.
Scenario 3: A public network is deployed along the railways and repeaters are installed in train carriages. With repeaters, the signal quality can be improved although the penetration loss is large.
Scenario 4: A dedicated network is deployed along the railways and repeaters are installed in carriages.
Current standard specifications have partly taken UE speeds up to 300 km/h into account, but only in the context of data demodulation, not for cell detection. With increased deployment of high speed train lines, increased number of UE users, and increased usage of bandwidth per user, dominating network operators are requesting improved UE performance and support for speeds exceeding 300 km/h. Future high speed trains are expected to travel at speeds above 500 km/h, e.g. the Superconducting Magnetic Levitation train (SCMaglev) to be deployed in Japan, where train sets have already reached 580 km/h in speed tests.
Apart from the relatively shortened time for detecting suitable neighbor cells for handover or cell reselection, high speed movement of the UE may also lead to significant Doppler shifts of the received radio signals. Such a Doppler shift forces the UE to increase its demodulation frequency when moving towards a cell (i.e. moving towards an antenna that defines a radio lobe of the cell), and decrease demodulation frequency when moving away from a cell, in order to maintain an acceptable receiver performance.
The Doppler shift can be expressed as:
      Δ    ⁢                  ⁢    f    =      f    ⁡          (                                                  1              -                              v                c                                                    1              +                              v                c                                                    -        1            )      
where c is the speed of light and v is the relative velocity of the UE towards the transmitting antenna. Referring to FIG. 1, an UE 101 is on a high speed train 103 on a railway track 104, connected to and moving away from cell A2 105 and quickly needs to detect cell B1 107 towards which the UE 101 is moving with a velocity vUE 109 of the train. According to current standard an antenna 111, 113 of a cell site can be as close as 2 m from the railway track 104, mainly motivated by that the wireless communication network would be integrated with the high-speed railway infrastructure. With an angle α between railway track 104 and a direction 106 to a cell antenna 113 and a UE velocity vUE, the relative velocity v towards the transmitting antenna giving rise to Doppler shift is v=vUE cos α.
The magnitude of the Doppler shift depends on the relative velocity of the UE 101 towards the transmitting antenna in a cell. Consequently, with transceivers located close to a path along which an UE is moving, i.e., a small angle between the trajectory of the UE and the line between the UE and the transmitting antenna, a substantial part of the UE velocity will transfer into a Doppler shift. Moreover there will be an abrupt change of sign of the Doppler shift when the UE passes the transmitting antenna and the smaller the angle, the more abrupt is the change from positive to negative Doppler shift.
Each radio propagation path may have its own Doppler shift, depending on how the radio waves travel between the transmitting antenna and the UE. In case of line-of-sight there is one dominant path, whereas in e.g. urban areas there is generally scatter (reflections) due to buildings to which the UE has a relative velocity, giving rise to multiple paths for the signal to propagate to the UE, each with a different Doppler shift. Since the received signal (in general) is the superposition of those paths, it gives rise to Doppler spread which degrades radio receiver performance by smearing out the signal in the frequency domain hence causing inter-carrier interference.
High-speed railway track sections are generally using dedicated platforms often elevated above the landscape or city beneath. Hence, there are few objects that can cause a significant Doppler spread; with cell sites located along the track line-of-sight will be dominating at least between the cell site and the train. Moreover, in built-up areas as well as when a train is approaching or passing stations the speed is generally restricted of concern for public safety and disturbing noise, and as a consequence the Doppler shift becomes small.
With regard to handover of a UE from a source cell to a target cell or, in scenarios where carrier aggregation is used, handover to a new primary cell, PCell, configuration of a new secondary cell, SCell, and configuration and activation of a new primary secondary cell, PSCell, is usually based on measurement reports from the UE, where the UE has been configured by the network node to send measurement reports periodically, at particular events, or a combination thereof. Such measurement reports typically contain physical cell identity, reference signal strength, RSRP, and reference signal quality, RSRQ, of the detected cells. Handovers can also be blind (i.e. no measurements performed on target carrier and/or cell) based on the network node having knowledge about coverage on other carriers and location of the UE. An example of this can be found in U.S. Pat. No. 8,892,103 entitled “Methods and nodes supporting cell change”.
The latency at a handover to a known (measured) PCell counted from reception of the handover command at the UE antenna until the UE carries out contention-free random access towards the target cell, can be up to 65 ms comprising 15 ms RRC procedure delay, 20 ms preparation time for the UE, and up to 30 ms latency for next physical random access channel, PRACH, occasion. One of the purposes with random access is to configure the UE with an appropriate timing advance value such that uplink transmissions by the UE are aligned with the subframe timing when received by the network node. Each random access attempt typically takes 20 ms hence in case the UE has to repeat the random access due to not getting response from the network node the time will be prolonged, but as a general figure one can assume 85 ms in total until the UE can resume communication in the target cell provided that the first attempt of random access is successful.
In the UE the preparation time is needed, e.g., for stopping processing and tearing down data structures and data memory associated with the source cell to release processing, memory and radio resources so they are available for the configuration to be used in the target cell. The reconfiguration may in general require re-partitioning of the data memories due to other bandwidth used in target cell, loading of new program code to support other transmission modes or radio access technology than in source cell.
However, there remain a number of challenges in relation to high-speed train scenarios in prior art. For example:                The UE experiences high Doppler shifts when passing an antenna node at a cell site, particularly when the site is close to the railway as in scenarios 1, 2 and 4 in RP-141849 summarized above. Such shifts also occur when the UE is handed over from one cell to another if the UE is moving in opposite directions to the cells. The abrupt shifts means that the UE will have to retune its radio before it can receive and transmit again, which is further prolonging a handover.        A UE that is between two antenna nodes, e.g. remote radio heads, in Scenarios 1, 2 and 4 in RP-141849 will receive the same signal from at least two directions, with opposite Doppler shifts. This gives rise to a Doppler spread also in the line-of-sight case, causing inter-carrier interference and degrades the receiver performance when both towers are received with about the same strength.        The abrupt change of sign for the Doppler shifts causes the frequency offset (discrepancy between UE demodulation frequency and the perceived carrier frequency) to be so large that it falls outside the capture range of the estimators. The UE risks tuning towards the wrong target with severely impacted performance and/or radio link failure as result.        Existing assumptions on network deployments for high-speed train scenarios assumes inter-cell distance of 300 m to 1000 m, which means that the UE changes or passes an antenna node at, e.g., a cell tower every 150 m to 500 m. This means that the UE will have to retune its receiver every 1.1 to 3.6 seconds when moving at a speed of 500 km/h. Each handover-related and/or Doppler-related interruption will have a significant impact on both system and UE throughputs.        Downlink, DL, and/or uplink, UL, Coordinated multi-point, CoMP, operation techniques can be deployed in both homogeneous and heterogeneous networks (see e.g. 3GPP TR 36.819 v11.1.0). But with high Doppler shifts and abrupt change of sign for the Doppler shifts it's impossible to use DUUL CoMP techniques to enhance base station and/or UE performance as it requires good frequency (equal or less than 300 Hz) and time tracking between Quasi-collocated base stations from multiple points.        